
Entropy-based test-time adaptation (EBTTA) methods can be interpreted from a clustering perspective, which provides insights to improve their performance by addressing challenges faced by clustering algorithms, such as sensitivity to initial assignments, nearest neighbor information, outliers, and batch size.


coremsg

improving-entropy-based-test-time-adaptation-by-leveraging-clustering-principles


Improving Entropy-Based Test-Time Adaptation by Leveraging Clustering Principles


title_rewrite


Incorporating active learning into the test-time adaptation setting can effectively mitigate distribution shifts and overcome catastrophic forgetting.


enhancing-test-time-adaptation-through-active-learning


Enhancing Test-Time Adaptation through Active Learning
